GPT 5.4 evaluated on our full benchmark suite

Gemini 3.1 Flash Lite evaluated across our benchmark suite

CaseLaw (v2) Benchmark Update

Qwen 3.5 Flash on Vals Index

Vibe Code Bench v1.1 Released

GPT 5.3 Codex Evaluated

Gemini 3.1 Pro evaluated across our benchmark suite

Gemini 3.1 Pro evaluated on Vals Index

MedCode Released: Can models support the medical billing process?

MedScribe Released: Can models support doctors with their administrative work?